Page MenuHomeAleph Objects Inc

Threads pulling out of extrusion
Closed, ResolvedPublic

Description

Just today we've had 5 frames, 2 from Cal and 3 that Kassandra caught while checking frames, where the threads will pull out of the top of the extrusion(HD-EX0062). Its always the top M5x10 that goes through the Z upper and into the top of the extrusion. Its really costly when the machine has made it to Cal and has to be reworked, but is still very much a pain to replace when its just a frame.

Related Objects

Event Timeline

robert created this task.May 21 2019, 4:03 PM
robert triaged this task as High priority.
robert created this object with edit policy "LulzBot Hardware Products (Project)".
Steven added a subscriber: Steven.May 22 2019, 7:16 AM

@kent are the threads on the extrusions being checked in QC? This likely needs to go to QC and supplier discussion.

@robert @logan @oliver @kent @MikeR @bigmansas @bowman
it sounds like these are being over tightened and that is stripping out the threads, we may want to check the calibration on the torque drivers being used in this stage, and ensure
that torque drivers are actually being used here

@tutley I've witnessed these being tightened with 2 fingers holding the driver and they strip out. They get checked with a 5in lb torque driver.

@tutley This could also be related to an issue were seeing with the extrusions look like they were drilled to large and then tapped with M5 threads. One think we need to do is perhaps have the frame assemblers be a bit more vigilant about what t-slot extrusions they are using.

@wolffchadd Tagging wolf in this ticket as well.

@robert yea those threads should be able to hold way more than 5 in*lbs torque. Based on @olivers comment, it seems like some of the extrusion is non conforming.

@Steven According to QC inspection sheet, they are check the whole size and threads but they only check 8% of them.

Steven added a subscriber: sherylb.May 22 2019, 7:40 AM

@sherylb @kent FYI, this looks to be leading to non-conforming materials from the supplier.

Kent will be in the building tomorrow and I will have him look at them. If we need to communicate to the vendor we will need the amount that are non conforming to send back and get a credit on.

The other day I had them stripping out when tightening them by hand, without putting a tool on them. It almost seems like only a couple threads on the screws are all that holds the extrusion the the uppers.

kent added a comment.May 22 2019, 8:03 AM

are the threads on the extrusions being checked in QC?

It looks like @oliver answered this already. We inspect 8%.

logan added a comment.May 22 2019, 8:22 AM

While I believe it is true that the extrusions are non-conforming having seen one myself this morning, we could stand to have better thread engagement on the top plate. The amount of thread engagement is reduced vs TAZ 6 because of the thicker top plate.
We could use HD-BT0158 M5x12 BHCS on top, and depending on what percentage becomes fallout, and whether or not that fallout rate would hinder production, this may be a potential interim solution. I have seen some of the top t-nuts strip out and not grab because the first 2 threads are gone.

kent added a comment.May 22 2019, 8:30 AM

@logan Is the HD-BT0158 the same longer fastener added to the frame for redgum recently for the chassis mounts and the thumbscrew leash? If so, then I think that's an easy fix since the hardware will already be stocked at the frame station.

logan added a comment.May 22 2019, 8:32 AM

@kent That is correct.

DaniAO assigned this task to kent.May 22 2019, 9:02 AM

i have already spoken to Kent about this he will handle this when he gets in tomorrow.

Just so the information is out there, I have just been informed that this issue is also present in the Taz 6 Y axis assembly.

Makes sense being that it is the same extrusion for both frame vertical and Y-Axis, thank you for making note of that. I believe the Y End plates are specified to the same thickness as TAZ 6 also, so definitely an issue with our double tapped extrusions on this delivery.

kent added a subscriber: samantha.May 23 2019, 8:17 AM

I checked some of the extrusions that we built the alpha units with and they had a minor diameter of 4.45 on average whereas the red tagged extrusions that we are seeing this issue with have a minor diameter of 4.75 on average so that explains it. I'll coordinate an inspection of the backstock and generate the NCMR. I will need to determine a threshold maximum minor diameter for these threads. @tutley @samantha do you have any recommendations? From the documents I've found, M5 x 0.8 threads should have a maximum of 4.33mm minor diameter, but I was thinking we could approve a variance up to 4.6mm without running into any issues.

@kent some of the M5 screws i measured the thread diameter of are around 4.85 on the low end, and 4.93 on the high end.
so i personally think 4.6 may be to high, as it doesnt leave much thread engagement across the range of screw diameters.
i can try and drill some holes to various sizes in a block of aluminum and then tap them and we can do a pullout and or strip out test to determine what will be the best range for us

We have had approximately 1965 more come into receiving. Can someone check a few of these as well on Tuesday to see if they have the same issue?

kent added a comment.May 24 2019, 3:26 PM

@sherylb yes I will plan on it.

kent added a comment.May 28 2019, 2:04 PM

I took a random sample of 40 extrusions from the pallet on the dock and measured the minor diameters with a pin gauge. Here are the results:

I think the parts in the new PO are going to work fine. It looks like if anything it's just a WIP issue.

sherylb added a subscriber: tomc.May 28 2019, 2:14 PM

@tomc @kent @wolffchadd please pull the ones from the dock to production if needed and move the others to a testing spot inside of QC and we will need to get on them as soon as possible. Thank you Kent for going through the test.

@kent yea those look pretty good. the largest measurement at 4.38 should leave plenty of thread engagement for the 5in*lbs we are applying.
I do agree with possibly looking at longer fasteners to allow more thread depth which will help prevent strip outs as well

kent lowered the priority of this task from High to Normal.May 29 2019, 8:17 AM
kent closed this task as Resolved.Jun 18 2019, 9:11 AM

This is just a NC materials issue, not a issue with the source and there are no changes to merge in, so this will be handled via the normal NC process.